Narrative Therapy in Context: When Does Telling the Story Help?
Edited with AI About this label
This version was adapted, rewritten, or refined with AI assistance from existing source material and/or human editorial input. Deep Emotional Work Academy retains editorial responsibility.
Review & Opinion
Telling a painful story can be part of a defined treatment, a conversation about identity, or an account intended to bear witness. Calling each of these “narrative therapy” hides differences that matter to the person receiving care.
The evidence supports some specific uses of narrative work. It also gives clear reasons to reject pressure to disclose and indiscriminate psychological debriefing. Neither finding settles the question for every approach that uses a person's story.
For practitioners and programme leads, the useful question is specific: which approach, for which person and purpose, under what conditions, and compared with what?
One word, several different approaches
Narrative therapy: identity, meaning and preferred stories
In the tradition developed by Michael White and David Epston, narrative therapy explores the stories through which people understand their lives. A central practice is externalising: examining a problem separately from the person's identity. Re-authoring conversations develop accounts of the person's values, responses, relationships and possibilities that a problem-dominated story has obscured. The person participates in deciding which meanings fit. Dulwich Centre's account of narrative approaches describes this tradition and its individual, group and community applications.
This is more specific than inviting somebody to talk, and it does not define therapy as a detailed retelling of the worst event. A discussion of identity or a person's response to adversity should not automatically be classified as exposure to traumatic memories.
There is outcome research, but its relevance depends on the question. A controlled study of 63 adults with moderate depression found symptom improvement in both narrative therapy and cognitive behavioural therapy. One depression measure favoured cognitive behavioural therapy; another broader outcome measure did not distinguish them. That supports a bounded discussion of depression treatment. It does not establish that unspecified narrative sessions treat post-traumatic stress disorder or are appropriate for everyone after a disaster.
Narrative Exposure Therapy: a defined trauma-focused treatment
Narrative Exposure Therapy (NET), developed by Maggie Schauer, Frank Neuner and Thomas Elbert, is a distinct approach. It places traumatic experiences within a person's life history and deliberately works with trauma memories. Its treatment manual describes a structured treatment, not an open invitation to disclose in any setting.
NET has a substantial research base relative to an undefined storytelling activity. A 2019 meta-analysis of 16 randomised trials involving 947 participants found better post-traumatic stress outcomes than inactive controls and active treatments without a trauma focus. It did not find superiority over the combined pool of active comparators. Most trials had an unclear risk of bias, and direct comparisons with other guideline-supported trauma-focused treatments were not available in that review. Improvement from before to after treatment is also a different result from outperforming another treatment.
The United Kingdom's National Institute for Health and Care Excellence guideline includes NET among individual trauma-focused cognitive behavioural treatments for adults with post-traumatic stress disorder (PTSD) or clinically important symptoms. Its recommendations specify a validated manual, trained practitioners and ongoing supervision. This is a treatment recommendation for an assessed clinical need, not a recommendation that everybody exposed to trauma should recount it.
Testimony therapy: a documented account and its meaning
Testimony therapy has a different history. An early report from Chile described a person's account of political repression being recorded and jointly revised into a written document. Later adaptations have combined testimony with culturally meaningful acknowledgment or ceremony.
A 2017 pilot randomised trial involving 88 Khmer Rouge torture survivors in Cambodia found reductions in PTSD, anxiety and depression symptoms with testimony therapy plus a culturally adapted ceremony compared with a control group. This is encouraging evidence for that combined intervention and population. It cannot isolate the effect of narration from the ceremony or establish the same benefit for another format.
NET draws partly on testimony therapy, but that shared history does not make the approaches interchangeable. Nor does testimony therapy establish that public disclosure is necessary for recovery.
A programme needs more than a method label
Mental health and psychosocial support (MHPSS) encompasses several levels of help. A non-governmental organisation (NGO) might describe therapeutic conversations, a named clinical treatment or a locally adapted activity using narrative language. The phrase “narrative therapy sessions” does not tell us which one it means.
The useful next step is to identify the manual or approach, purpose, participant selection, delivery format and staff preparation. Community delivery describes a setting and a way of reaching people. It does not establish which intervention they receive or whether it fits their needs.
The Inter-Agency Standing Committee's emergency guidelines place community support, focused help and specialised care within a layered response. These are different functions. An activity's name cannot substitute for a description of its function and delivery.
The strongest warning concerns pressure and indiscriminate debriefing
The World Health Organization's (WHO) psychological first aid guidance makes room for listening while explicitly rejecting pressure to recount experiences. The Academy's Psychological First Aid in Context examines that first-response role in more detail.
This matters because listening to somebody who wants to speak is different from organising a session on the assumption that everyone needs to tell the story.
Psychological debriefing is a specific early intervention intended to process a recent event through a structured account of experiences and reactions. A Cochrane review found that single-session individual debriefing did not prevent PTSD or reduce distress. One trial found increased PTSD risk at one year. WHO recommends against psychological debriefing to prevent post-traumatic stress, anxiety or depressive symptoms after recent exposure.
The limits matter. Those findings concern an early preventive intervention, principally single-session individual debriefing in the review. They do not show that all trauma narration, all group support or established trauma-focused treatments are harmful. Equally, evidence for a defined treatment cannot justify an improvised disclosure session.
Safety and preparation matter. A compulsory preliminary phase is another claim.
There are two different questions here: whether care needs safety, assessment and preparation, and whether everyone must complete a separate stabilization phase before trauma-focused treatment.
A randomised trial in adults with childhood-abuse-related PTSD compared a skills-training phase followed by eye movement desensitisation and reprocessing (EMDR) with immediate EMDR. Both groups improved, without significant differences between the treatment conditions. The study's selection criteria limit the conclusion: people needing crisis intervention for acute suicidality and people experiencing ongoing physical or sexual abuse were among those excluded. This was also an EMDR comparison, not a test of every narrative approach.
A 2025 systematic review of four trials involving 356 adults with childhood-trauma-related PTSD found limited added value for phase-based approaches within the available evidence. A 2026 review of interventions for complex PTSD found no significant differences across most outcomes. Some subgroup comparisons favoured multi-phase approaches or particular emotion-regulation outcomes, but the evidence was limited by small numbers of trials, varied measures and no pooled follow-up. Taken together, these findings support attention to the person's needs, not a universal sequence. They do not justify imposing a separate preliminary phase on everyone or removing preparation from care.
Our position is that support should widen access to appropriate care. Stabilization must not become an indefinite waiting room or a compulsory gateway to effective treatment. The same distinction informs the Academy's discussion of Stabilization Groups in Context.
Conditions that change the decision
Before a programme deliberately invites detailed trauma narration, its responsible team should be able to explain the following. These are programme-review questions drawn from the guidance and research above, not a treatment protocol or a self-assessment for survivors.
Purpose and fit. What need is being addressed, and why is this approach appropriate for this person? Basic support, treatment of persistent symptoms and the creation of testimony are different purposes. A general wish to improve well-being is not enough to select a trauma-processing procedure.
Choice and control. Does the person understand the proposed work, available alternatives and the possibility of distress? Can they decline disclosure, pause or withdraw without losing unrelated support? Consent to receive help should not be treated as consent to tell a detailed story. WHO's first-aid guidance places choice and freedom from pressure at the centre of support.
Safety and practical readiness. What immediate protection or care needs take priority? What would make the session inappropriate today? The aim is a workable care plan, not a demand that the person's whole life become stable before help is available.
Competence and supervision. Who can deliver this particular intervention, recognise problems and obtain help? Training has to match the task. A randomised refugee-settlement trial studied delivery by trained lay counsellors. That shows why the distinction is not simply specialist versus non-specialist: a defined intervention and its training arrangements matter. It gives no authority to transfer trauma processing to unprepared volunteers.
Format and privacy. Is the work individual or group-based, and what evidence supports that format? Individual-treatment findings do not automatically validate a group adaptation. A programme should explain who hears the account, the limits of confidentiality, and whether anything is recorded or shared. The Inter-Agency Standing Committee guidance addresses informed consent and protection of confidential information. Our practical implication is straightforward: inviting a detailed account before other participants adds disclosure decisions that private treatment does not have.
Monitoring and continuity. Who follows up if the person becomes more distressed, stops attending or needs another service? What outcomes, deterioration and complaints will the programme examine? Attendance and immediate appreciation alone cannot establish lasting benefit. Follow-up should test whether the intended benefit occurs and whether the chosen format is causing avoidable problems.
These conditions support scrutiny. They cannot turn an unsuitable method into an effective one, and a completed checklist cannot guarantee safety.
Choose the response to match the need
For immediate distress after a crisis, practical assistance, protection, a willing listener and connection with support may be the relevant first response. A detailed trauma account is not the price of receiving that help.
For persistent, clinically important trauma symptoms, assessment and access to a suitable evidence-based treatment are the relevant comparison. Avoiding every discussion of traumatic memory can also prevent someone from receiving an effective treatment they want. For identity and meaning questions, narrative therapy should be evaluated on its own purpose and evidence. For testimony work, the person must understand the intended use and audience of the account.
Our critical conclusion is therefore specific: “telling the story” is too imprecise to serve as a programme rationale. The team needs to name the intervention, explain its fit and show the conditions under which it is delivered. The participant needs a real choice.
Before adopting or continuing narrative sessions, compare that actual approach with the alternatives for the same need. If the protocol is unknown, establish it before judging the service. If the conditions are missing, address them before inviting detailed disclosure. A critical review should protect both the freedom to speak and the freedom to receive help without speaking.
Professional scope
Matthias Behrends contributes as an educator and instructor. All activities in instructor-led institutional training require the involvement of a licensed clinician and a designated supervisor.
